Order ready-to-submit essays. No Plagiarism Guarantee!
Note: All our papers are written from scratch by human writers to ensure authenticity and originality.
Research analysis
Check your essay before you submit. See exactly what your professor sees.
See your AI and plagiarism results before your instructor does.Get the exact same report your professor uses. Trusted by 50,000+ students worldwide.
Give an annotated summary (max 100 words) for each one: (bullet point for answers
What is the main hypothesis?
Why is this research important?
How was the research question studied (provide paradigm)?
Observation, survey, case study, experiment, etc.
What were the variables in the study?
What statistical tests were conducted?
What was the key finding of the research?
What are the conclusions and limitations?
Why is this work relevant for your group work?
BRIEF REPORT
Concreteness effects in bilingual and monolingual
word learning
Margarita Kaushanskaya & Katrina Rechtzigel
Published online: 26 May 2012
Psychonomic Society, Inc. 2012
Abstract Previous studies have demonstrated that bilingualism
can facilitate novel-word learning. However, the mechanisms
behind this bilingual advantage remain unknown. Here,
we examined whether bilinguals may be more sensitive to
semantic information associated with the novel words. To that
end, we manipulated the concreteness of the referent in the
word-learning paradigm, since concrete words have been
shown to activate the semantic system more robustly than
abstract words do. The results revealed that the bilingual
advantage was stronger for novel words learned in association
with concrete rather than abstract referents. These findings
suggest that bilingual advantages for word learning may be
rooted, at least in part, in bilinguals’ greater sensitivity to
semantic information during learning.
Keywords Human memory and learning . Bilingualism .
Word meaning . Phonology . Semantics
Previous studies have suggested that bilingualism can facilitate
performance on word-learning tasks (e.g., Kaushanskaya
& Marian, 2009; Papagno & Vallar; 1995; Van Hell & Mahn,
1997). However, the mechanisms that underlie the effects of
bilingualism on learning are currently unknown. One early
suggestion was that the bilingual advantage for word learning
was based in bilinguals’ more efficient phonological memory
system (e.g., Papagno & Vallar, 1995). However, in recent
work we have shown that bilingual advantages for novel word
learning are maintained when bilingual and monolingual participants
are matched precisely on their phonological memory
capacity (Kaushanskaya, in press). Therefore, it is likely that
bilingual experience affects the learning process itself, rather
than (or in addition to) the cognitive workspace (i.e., the
working memory) within which learning takes place.
Although learning a word involves encoding both its phonological
form and its meaning, and although learning can
occur in a number of different ways, previous studies that
examined the effects of bilingualism on word learning have
typically used the paired-associates learning paradigm (rather
than other types of word-learning paradigms) and probed for
participants’ ability to retrieve the native-language translations
as the index of learning (e.g., Kaushanskaya, in press;
Kaushanskaya & Marian, 2009; Van Hell & Mahn, 1997). A
promising clue to how bilingualism may influence this particular
word-learning process was yielded by analyses of wordlearning
errors made by bilingual versus monolingual learners
(Kaushanskaya & Rechtzigel, 2012). In this previous study,
participants were taught to associate novel words with English
translations, and at testing, they were asked to produce the
English translations when cued with the novel words. While a
general pattern of bilingual advantages was uncovered, more
interesting findings were observed when translation errors
were analyzed. Errors were coded into two categories:
sound-based errors (in which participants confused two
similar-sounding novel words) and meaning-based errors (in
which participants confused two English translations that
shared a semantic category or that were associatively related
to each other). The analyses of errors were conducted on the
proportion data (e.g., on the number of sound-based errors out
of the total number of errors) to correct for the fact that
bilinguals made fewer errors overall than did monolinguals.
When analyzed in this way, bilinguals made more meaningbased
errors than did monolinguals. These results were
Electronic supplementary material The online version of this article
(doi:10.3758/s13423-012-0271-5) contains supplementary material,
which is available to authorized users.
M. Kaushanskaya () : K. Rechtzigel
Department of Communicative Disorders,
University of Wisconsin–Madison,
1975 Willow Drive,
Madison, WI 53706, USA
e-mail: kaushanskaya@wisc.edu
Psychon Bull Rev (2012) 19:935–941
DOI 10.3758/s13423-012-0271-5
interpreted to suggest that given the same task parameters and
stimuli, bilinguals may be able to encode novel words more
deeply (i.e., to the semantic level) than do monolinguals. The
goal of the present study was to experimentally assess whether
this was the case. We asked: Are bilinguals more sensitive to
the semantic information associated with novel words during
learning than are monolinguals?
In order to examine whether bilinguals and monolinguals
differ in how they process semantic information during learning,
we contrasted the learning of novel words in association
with concrete versus abstract referents. In memory tasks, concreteness
effects have been broadly confirmed (e.g., Hamilton
& Rajaram, 2001; Miller & Roodenrys, 2009; Romani,
McAlpine, & Martin, 2008; Walker & Hulme, 1999) using
both paired-associates learning (e.g., Gee, Nelson, &
Krawczyk, 1999) and novel-word learning paradigms (e.g.,
De Groot & Keijzer, 2000). Although the roots of concreteness
effects continue to be debated, nearly all memory/learning
studies that contrast concrete versus abstract words have
revealed better performance on concrete words (e.g., De
Groot & Keijzer, 2000; Hamilton & Rajaram, 2001; Miller &
Roodenrys, 2009; Romani et al., 2008; Walker & Hulme,
1999), and all of the theoretical explanations for why concrete
words are advantaged during processing entail semantic (rather
than phonological) differences between concrete and abstract
words. Thus, some studies have indicated that concrete words
are processed by both the verbal and the image-based systems,
while abstract words activate the verbal system only (Paivio,
1986, 1991; Paivio, Walsh, & Bons, 1994). Others have argued
that concreteness effects are due to richer semantic representations
(which may involve a wider network of prior contextual
knowledge) for concrete words (e.g., De Groot, 1989; Grondin,
Lupker, & McRae, 2009; Schwanenflugel & Shoben, 1983).
For the purposes of the present study, we assumed that the ease
associated with retaining concrete words is rooted in semantic
factors. We therefore hypothesized that if bilinguals were more
sensitive to semantic information during learning than monolinguals,
the bilingual advantage for novel-word learning
would be especially strong for concrete words.
The comparison of bilinguals and monolinguals on learning
tasks contrasting the learning of novel words in association
with concrete versus abstract referents also enabled us
to examine the strength of concreteness effects in the two
groups of learners. Theories of lexical processing in bilinguals
generally posit that the semantic representations of
translation equivalents in a bilingual’s two languages overlap
(e.g., Kroll & Stewart, 1994). However, there have also
been suggestions that different types of words overlap in the
bilingual semantic system to different degrees. Specifically,
the distributed feature model (De Groot, 1992) posits that
concrete words are more likely to share semantic features
across bilinguals’ two languages than are abstract words, and
some behavioral evidence has suggested that concrete
translation pairs across the bilingual’s two languages may
share a larger semantic overlap than do abstract translation
pairs (e.g., Jin, 1990; Paivio & Desrochers, 1980; Van Hell &
De Groot, 1998). As a result, bilinguals tend to translate
concrete words more quickly than abstract words (De Groot,
Dannenburg, & Van Hell, 1994; De Groot & Poot, 1997; but
see Tokowicz & Kroll, 2007, for an alternative explanation).
Using the distributed feature model as the framework, it is
possible to generate specific predictions with regard to the
strengths of the concreteness effects in bilingual versus monolingual
learning. In monolinguals, presentation of a concrete
word activates a wider lexical–semantic network than does the
presentation of an abstract word (e.g., De Groot, 1989;
Grondin et al., 2009; Schwanenflugel & Shoben, 1983), and
the same is likely true for bilingual speakers. However, presentation
of a concrete item in one of the bilinguals’ two
languages is likely to activate a wider lexical–semantic network
than does the presentation of the same item to a monolingual,
because in a bilingual, activation will include both the
language in which the word was presented and the other
language. Conversely, presentation of an abstract item in one
of the bilingual’s two languages is likely to activate a network
similar to that in a monolingual speaker, since abstract translation
equivalents do not overlap in the semantic network of a
bilingual speaker to the same extent as concrete translation
equivalents. If activation of the lexical–semantic system is
stronger for concrete than for abstract words, and if bilinguals’
lexical–semantic system is more robustly activated in response
to the concrete words, then concreteness effects should
be larger for bilingual than for monolingual learners.
Method
Participants
A group of 44 participants completed the study. Of these, 22
participants were monolingual native speakers of English who
reported no significant knowledge of a second language (defined
as self-ratings of second-language [L2] speaking proficiency
of 2 or less on a scale from 0 [no knowledge] to 10
[native-like knowledge]
), and 22 were English–Spanish bilinguals.
The bilingual participants were carefully screened to
ensure the following: English as the native language, selfrated
average proficiency in Spanish of 7 or above (averaged
across speaking and understanding ratings), and at least one
immersion experience of at least 2 months in length in a
Spanish-speaking country or family. Upon selection, bilingual
participants were also administered a standardized receptive
vocabulary test in Spanish in order to ensure adequate levels of
Spanish knowledge. The demographic characteristics of the
bilingual participants are reported in Table 1. All the demographic
and language experience information was collected
936 Psychon Bull Rev (2012) 19:935–941
using the Language Experience and Proficiency Questionnaire
(Marian, Blumenfeld, & Kaushanskaya, 2007).
Although bilingual and monolingual participants were
recruited from the same college-student population of
the University of Wisconsin–Madison, the requirement
for immersion experience in the bilingual group resulted
in slightly unbalanced groups (see Table 2). Specifically,
bilinguals were on average 1.87 years older than the
monolinguals, with 1.64 more years of education. The
two groups were matched in English receptive vocabulary
skills and in verbal working memory skills. All
participants were screened for nonverbal intelligence to
ensure within-normal cognitive skills. Raw scores on the
Kaufman Brief Intelligence Test–II were comparable
between the two groups (p 0 .17). However, when raw
scores were transformed into standard scores, the age
difference between the two groups resulted in the hypercorrection
of the bilinguals’ scores. The result of this
hypercorrection was a significant difference between
bilinguals’ and monolinguals’ standard IQ scores.
Materials
Two lists of 12 novel words were selected from the Gupta et
al. (2004) nonword database. All novel words were two
syllables in length and followed the sound patterns of
English (e.g., gapume, botefe). None of the novel words were
real words in Spanish. The stimuli were matched across the
two lists on syllable structure, in that all novel words were
consonant–vowel–consonant–vowel–consonant pseudowords.
The novel words across the two lists did not differ in
duration [List A, M 0 999.17 ms, SD 0 69.05; List B,
M 0 1,015.67 ms, SD 0 81.00; t(11) 0 0.62, p 0 .55], phonotactic
probability [List A 0 0.23, SD 0 0.04; List B 0 0.23,
SD 0 0.04; t(11) 0 0.33, p 0 .75], or biphone frequency [List
A 0 0.007, SD 0 0.003; List B 0 0.007, SD 0 0.04; t(11) 0 0.62,
p 0 .55]. The phonotactic and biphone frequency values were
calculated using the phonotactic probability calculator
(Vitevitch & Luce, 2004). The two lists of nonword stimuli
can be found in supplementary materials for this article.
Each novel word was paired with an English word. One
list of novel words was paired with concrete English nouns
(e.g., daisy, helmet), and the other list of novel words was
paired with abstract English nouns (e.g., virtue, sorrow).
The pairing of novel-word lists with English noun lists
(concrete vs. abstract) was counterbalanced across participants.
Concreteness was determined using the MRC lexical
database. Only nouns with concreteness ratings above 500
(on a scale from 100 to 700) were selected for the concrete
list, and only nouns with concreteness ratings below 350
were selected for the abstract lists. The two lists of English
words (concrete vs. abstract) differed significantly in their
concreteness ratings (concrete, M 0 588.67, SD 0 24.60;
abstract, M 0 279.42, SD 0 31.74) [t(11) 0 29.81, p <>
However, the two lists of English words were matched on
syllable length, frequency of use [concrete, M 0 36.16,
SD 0 40.79; abstract, M 0 37.50, SD 0 43.39; t(11) 0 1.07,
p 0 .31], phonotactic probability [concrete, M 0 0.28,
SD 0 0.10; abstract, M 0 0.30, SD 0 0.13; t(11) 0 0.45,
p 0 .66], biphone frequency [concrete, M 0 0.02, SD 0 0.01;
abstract, M 0 0.02, SD 0 0.02; t(11) 0 0.49, p 0 .63], and number
of lexical neighbors [concrete, M 0 0.75, SD 0 1.14; abstract,
M 0 1.00, SD 0 2.59; t(11) 0 0.29, p 0 .77]. Frequency of use
(per million) and neighborhood density values were calculated
using CELEX (Baayen, Piepenbrock, & Gulikers, 1995).
The novel words and the English words were recorded by
two different female speakers, both of whom were native
speakers of American English. All stimuli were recorded in
a soundproof booth at a 44-kH sampling rate and were
normalized to 70-dB amplitude using Praat (Boersma &
Weenink, 2007).
Table 1 Spanish (L2) acquisition and L2 use data for bilingual participants
(means and SDs)
L2 Data Mean (SD)
Test de Vocabulario en Imagenes de Peabody
(standard score)
113.30 (4.10)
Age of L2 acquisition 9.50 (4.80)
Degree of current L2 exposure 8.70 % (6.50)
Self-rated L2 proficiency: Speaking (0–10 scale) 7.50 (0.90)
Self-rated L2 proficiency:
Understanding (0–10 scale)
8.10 (1.10)
Years of immersion in a Spanish-speaking country 1.50 (3.60)
Table 2 Monolingual and bilingual participants data (means and SDs)
Monolinguals
(n022)
Bilinguals
(n022)
t
Value
Age 20.07 (1.0) 21.94 (1.8) 4.34
Years of education 14.6 (1.0) 16.3 (1.3) 4.37*
Peabody Picture
Vocabulary Test–III
(standard scores)
111.5 (4.4) 116.6 (13.3) 2.57
Woodcock Johnson Tests
of Cognitive Abilities–II
Numbers Reversed
(standard scores)
109.8 (12.7) 113.8 (10.8) 1.11
Kaufman Brief
Intelligence Test–II
Visual Matrices
(raw scores)
38.77 (0.72) 40.45 (0.95) 1.41
Kaufman Brief
Intelligence Test–II
Visual Matrices
(standard scores)
103.45 (2.13) 111.55 (2.80) 2.27*
- Significant differences between monolinguals and bilinguals, p <>
Psychon Bull Rev (2012) 19:935–941 937
Procedure
Each participant completed the entire experiment in a single
testing session. During the learning phase, novel words and
their English translations were presented over computer
speakers in a soundproof booth. The novel word was always
presented first, followed by the English word. The presentation
of concrete and abstract pairs was intermixed. Each
pair was presented twice during the learning phase, with the
order of presentation randomized for each participant. The
interstimulus interval between the presentation of the novel
word and its English translation was set to 750 ms, and the
interval between presentations of the pairs was set to 3 s.
Participants were instructed to memorize the association
between the novel words and their English translations.
During the testing phase, participants were asked to produce
the correct English translations for the novel words.
Such translation-like tasks are common measures of associative
learning (e.g., De Groot & Keijzer, 2000; Paivio &
Yuille, 1969) and have frequently been used in previous
studies of novel-word learning with both bilinguals and
monolinguals (e.g., Kaushanskaya & Marian, 2009; Van
Hell & Mahn, 1997). Each novel word was played over
speakers, and the participants were instructed to produce
the corresponding English translation into a microphone.
The responses were recorded as sound files for later coding
and analyses.
During the standardized testing phase, standardized
language and intelligence tests were administered to
each participant, including an English receptive vocabulary
test (Peabody Picture Vocabulary Test–III), a nonverbal
IQ test (Kaufman Brief Intelligence Test, Visual
Matrices subtest), and a verbal working memory test
(Numbers Reversed subtest of the Woodcock Johnson
Tests of Cognitive Abilities–II). A measure of Spanish receptive
vocabulary was administered to the bilingual speakers
(Test de Vocabulario en Imágenes de Peabody).
Analyses
The production accuracy data (proportions correct) were
normally distributed (Kolmogorov–Smirnov statistics
< 0.15,=”” p=””>.1), and therefore were analyzed using a 2 ×
2 ANOVA, with group (monolingual vs. bilingual) and concreteness
(concrete vs. abstract) as independent variables.
Both by-subjects (F1) and by-items (F2) analyses are reported.
Production reaction time (RT) data were also analyzed using a
2 × 2 ANOVA. RTs were measured from the onset of the novel
word to the participant’s buttonpress after production of the
English translation. Only RTs for the correctly produced translations
were included in the calculations of mean RTs. Years of
education and standard scores on the nonverbal IQ measure
were covaried out in all of the cross-group by-subjects
analyses. Age was not covaried out, because it correlated
highly with years of education.
Results
Accuracy analyses
The proportion correct data across groups and conditions are
presented in Fig. 1. A 2 × 2 ANOVA revealed a main effect
of concreteness [F1(1, 41) 0 4.26, MSE 0 0.05, p <>
?p
2 0 .10; F2(1, 22) 0 17.37, MSE 0 0.23, p <>
?p
2 0 .44], with concrete referents (M 0 .22, SE 0 .02)
retrieved more accurately than abstract referents (M 0 .11,
SE 0 .02). The main effect of group was not significant in the
by-subjects analysis [F1(1, 41) 0 2.38, MSE 0 0.02, p 0 .13,
?p
2 0 .09], but was significant in the by-items analysis
[F2(1, 22) 0 6.32, MSE 0 0.04, p < .05,=””>
2 0 .22]. Bilinguals
(M 0 .19, SE 0 .02) tended to outperform monolinguals
(M 0 .13, SE 0 .03). Crucially, the interaction between concreteness
and group was significant [F1(1, 41) 0 5.01, p < .05,=”” mse=””>
0.06, ?p
2 0 .12; F2(1, 22) 0 5.55, p < .05,=”” mse=”” 0=”” 0.04,=””>
2 0 .11].
Follow-up univariate ANOVAs with group as the independent
variable revealed that bilingual participants were more
accurate than monolingual participants when retrieving concrete
referents for the newly learned words [F1(1, 42) 0 5.54,
MSE 0 0.10, p < .05,=””>
2 0 .12; F2(1, 11) 0 4.91,
MSE 0 0.03, p < .05,=””>
2 0 .31]. However, bilinguals and
monolinguals were similarly accurate when retrieving abstract
referents for the newly learned words [F1(1, 42) 0 0.19,
MSE 0 0.01, p 0 .67, ?p
2 0 .02; F2(1, 11) 0 1.70, MSE 0 0.01,
p 0 .22, ?p
2 0 .13].
Follow-up repeated measures ANOVAs with concreteness
as the independent variable were used to examine the
strength of the concreteness effects within each group separately.
These analyses revealed a significant concreteness
effect in the bilingual group [F1(1, 21) 0 21.51,
MSE 0 0.27, p < .001,=””>
2 0 .51; F2(1, 22) 0 10.71,
0
0.1
0.2
0.3
0.4
0.5 Abstract Referents
Concrete Referents
Proportion Correct
Monolinguals Bilinguals
*
*
*
Fig. 1 Mean monolingual and bilingual translation accuracy (proportions
correct) in the concrete-referent and abstract-referent conditions.
The error bars represent standard deviations. *
p <>
938 Psychon Bull Rev (2012) 19:935–941
MSE 0 0.15, p < .01,=””>
2 0 .33] and in the monolingual
group [F1(1, 21) 0 8.57, MSE 0 0.11, p < .01,=””>
2 0 .29;
F2(1, 22) 0 14.27, MSE 0 0.09, p < .01,=””>
2 0 .39]. However,
comparisons of the effect sizes in the by-subjects data
revealed that the effect of concreteness was significantly
stronger in the bilingual group than in the monolingual group,
z score 0 12.96, p <>
RT analyses
A 2 × 2 ANOVA revealed a marginally significant effect of
concreteness in the by-subjects analyses [F1(1, 23) 0 4.06,
MSE 0 40,533,371.77, p 0 .056, ?p
2 0 .15], but not in the
by-items analyses [F2(1, 18) 0 0.00, MSE 0 0.13, p 0 .10,
?p
2 < .001].=”” there=”” was=”” a=”” tendency=”” for=”” concrete=”” referents=”” (m=””>
5,108.58, SE 0 677.95) to be retrieved more quickly than
abstract referents (M 0 6,922.43, SE 0 514.37). The main
effect of group was not significant [F1(1, 23) 0 0.29, MSE 0
2,242,493.76, p 0 .60, ?p
2 0 .01; F2(1, 18) 0 0.002, MSE 0
9,709.26, p 0 .96, ?p
2 < .001].=”” moreover,=”” the=””>
between concreteness and group was also not significant
[F1(1, 23) 0 2.80, MSE 0 27,903,646.31, p 0 .11, ?p
2 0 .11;
F2(1, 18) 0 2.75, MSE 0 11,677,855.14, p 0 .12, ?p
2 0 .13].
Discussion
In the present study, we contrasted the learning of novel words
in association with concrete versus abstract English translations
in order to test the hypothesis that the bilingual advantages
for novel-word learning may be based, at least in part, on
bilinguals’ sensitivity to the semantic information associated
with the novel words. Prior extensive work on concreteness
effects strongly supports semantic differences between abstract
and concrete words (De Groot, 1989; Grondin et al.,
2009; Paivio, 1986, 1991; Paivio et al., 1994; Schwanenflugel
& Shoben, 1983), with concrete words activating a richer
network of semantic information (e.g., mental imagery or a
wider network of contextual information) than abstract words.
Our finding of a stronger bilingual advantage for concrete than
for abstract novel words suggests that the effects of bilingualism
on word learning are more likely to emerge when semantic
information associated with the novel words is more
accessible. The lack of significant RT differences between
bilinguals and monolinguals on the translation task indicates
that bilinguals’ more accurate translation performance was not
a result of accuracy–RT trade-offs or of bilinguals’ strategic
allocation of time to the translation task.
What may explain the bilinguals’ superior learning of
concrete words? Consideration of concreteness effects in the
monolingual group versus the bilingual group is helpful in
informing this question. Although a main effect of concreteness
was found, it is apparent that concreteness effects were
stronger in the bilingual group than in the monolingual group.
The difference between the concrete and abstract conditions in
the bilingual group (mean difference 0 .17) was more than
twice the size of the difference in the monolingual group
(mean difference 0 .08). Previous work on the representations
of concrete versus abstract words in the bilingual lexical–
semantic system suggests that concrete translation pairs across
bilinguals’ two languages may share a larger semantic overlap
than do abstract translation pairs (e.g., De Groot, 1992; Jin,
1990; Paivio & Desrochers, 1980; Van Hell & De Groot,
1998). Therefore, it is possible that concrete words cause a
wider activation of the bilingual lexical–semantic system (vs.
the monolingual system), thus yielding a stronger concreteness
effect in bilinguals than in monolinguals. It may be that
the same mechanism—namely, the structure of the bilingual
(vs. the monolingual) lexical–semantic system—is what drove
the stronger bilingual advantage for the concrete than for the
abstract words. The English translations would have activated
a wider lexical–semantic network in bilinguals than in monolinguals,
since in bilinguals the activation would have involved
both the activation of the English lexical–semantic
network and the overlapping Spanish lexical–semantic network.
This would result in more robust lexical–semantic
activation in response to concrete words in bilinguals, thus
yielding stronger learning. Conversely, abstract English translations
would have activated similar lexical–semantic networks
in bilinguals and in monolinguals, yielding comparable learning
profiles in the two groups of learners.
Thus, it may be that sensitivity to semantic information
during learning is a general property of the bilingual processing
system, where through exposure to two languages,
the bilingual’s semantic system becomes more interactive,
especially when confronted with concrete stimuli. That is,
bilingual advantages for the learning of concrete words may
stem not from different learning mechanisms in bilinguals
versus monolinguals, but from the higher levels of semantic
activation in the bilingual versus the monolingual lexical–
semantic system, as a result of the availability of two languages.
Thus, bilingual advantages for novel-word learning
may be a natural outcome of how words are stored and
processed in the bilingual versus the monolingual lexical–
semantic system. This explanation of bilingual advantages
on word-learning tasks is theoretically akin to Gollan and
colleagues’ (Gollan & Acenas, 2004; Gollan, Montoya,
Fennema-Notestine, & Morris, 2005; Gollan & Silverberg,
2001) explanation of bilingual disadvantages on wordprocessing
tasks. Gollan et al. have suggested that bilinguals’
reduced lexical retrieval abilities are an outcome not
of fundamental differences in language processing between
bilinguals and monolinguals, but of the fact that bilinguals,
by virtue of distributed lexical exposure, have weaker links
between lexical and semantic representations. Here, it is possible
that bilinguals’ more efficient lexical learning is an
Psychon Bull Rev (2012) 19:935–941 939
outcome of a more resonant and interactive semantic system
in bilinguals as a result of cross-linguistic coactivation.
The stronger effect of bilingualism on the learning of
concrete (rather than abstract) novel words, although suggestive
of a mechanism that underlies bilingual advantages
for word-learning tasks, leaves open a number of additional
possibilities as to the identity of this mechanism. For example,
it is possible that bilinguals may benefit from concrete
referents because they recognize that focusing on the words’
meanings would be a successful strategy on this particular
type of learning task, and thus they choose to consciously
employ this strategy. It may be feasible in future studies to
examine whether bilinguals’ sensitivity to semantic information
is an outcome of strategy use by instructing learners
to use semantic versus rote (phonological) strategies when
encoding novel words. If bilinguals are able to strategically
allocate attention to the semantic information associated
with the novel words, they should be particularly skilled at
utilizing meaning-based strategies at learning, independent
of the semantic content of the words (concrete or abstract).
If, on the other hand, the interactivity of the bilingual
lexical–semantic system is at the root of the bilingual advantages
for learning concrete words, then manipulating learning
strategy should not influence the patterns observed in
the present study. That is, independent of how learners are
instructed to learn, bilinguals would be expected to outperform
monolinguals on the concrete but not on the abstract
items.
The interpretation of the bilingual advantage for word
learning observed in the present study must be qualified by
the following considerations. First, this word-learning task
is well practiced by experienced learners. In fact, Van Hell
and Mahn (1997) demonstrated that experienced language
learners (i.e., multilingual learners with large amounts of
classroom-based language-learning experience) were especially
successful at learning novel words via a pairedassociates-learning-like
method. Therefore, the finding of a
bilingual advantage in the present study may apply only to
bilinguals with vast amounts of L2 experience and to learning
tasks of this particular kind. It is interesting to note that
although our strict inclusion criteria for bilingual participants
resulted in a homogeneous group of bilinguals with
respect to their L2 proficiency, there was variability in the
bilingual group with respect to the age at which the L2 was
acquired. When bilingual participants’ word-learning data
were correlated with L2 age-of-acquisition data, inverse
correlations were observed between age of acquisition and
translation accuracy. The correlation was significant for
abstract words (r 0 –.45, p < .05),=”” but=”” not=”” for=”” concrete=””>
(r 0 –.18, p 0 .42). The fact that earlier exposure to the L2
was associated with better word-learning performance indicates
that robust exposure to an L2 is likely to be crucial for
the positive effects of bilingualism to emerge on learning
tasks of the kind used in the present study. Second, the
finding that the bilingual advantage for word-learning tasks
is qualified by concreteness may be specific to retention
measures that probe for the meanings associated with the
novel words, as was the case in the present study. Therefore,
it will be important to test both the retention of the novel
word forms and of their meanings in the future. Third,
although the translation accuracy rates in the present study
were quite similar to those in other published accuracy data
on translation-like tasks administered immediately after
learning (e.g., Kaushanskaya, in press), they are suggestive
of the high levels of difficulty associated with this particular
learning task. Therefore, it is possible that bilingual advantages
on learning tasks are especially likely to be observed
when tasks are quite challenging.
In conclusion, the goal of the present work was to inform
the search for the mechanisms that may underlie the effects
of bilingualism on learning. We found that the bilingual
advantage was stronger for concrete than for abstract novel
words, and we ascribe this advantage to bilinguals’ greater
sensitivity to the semantic information associated with the
novel words. It may be that the bilingual lexical–semantic
system is more robustly activated in response to semantically
rich material than is the monolingual lexical–semantic
system, yielding stronger concreteness effects and better
retention of concrete words in bilingual versus monolingual
learners.
Author note This research was supported in part by a University of
Wisconsin–Madison Graduate School Research Committee WARF
Grant to M.K. The authors thank Matt Goldrick, Nicolas Dumay, two
anonymous reviewers, and the members of the Language Acquisition
and Bilingualism Lab for their helpful suggestions on this manuscript,
and the Study Abroad Office of the University of Wisconsin-Madison
for their help with participant recruitment.
References
Baayen, R. H., Piepenbrock, R., & Gulikers L. (1995). The CELEX
lexical database (Release 2). Philadelphia, PA: Linguistic Data
Consortium, University of Pennsylvania.
Boersma, P., & Weenink, D. (2007). Praat: Doing phonetics by computer
(Version 4.6) [Computer program]. Retrieved from www.
praat.org.
De Groot, A. M. B. (1989). Representation aspects of word imageability
and word frequency as assessed through word association.
Journal of Experimental Psychology: Learning, Memory, and
Cognition, 15, 824–845.
De Groot, A. M. B. (1992). Determinants of word translation. Journal
of Experimental Psychology: Learning, Memory, and Cognition,
18, 1001–1018.
De Groot, A. M. B., & Keijzer, R. (2000). What is hard to learn is easy
to forget: The roles of word concreteness, cognate status, and
word frequency in foreign-language vocabulary learning and forgetting.
Language Learning, 50, 1–56.
940 Psychon Bull Rev (2012) 19:935–941
De Groot, A. M. B., & Poot, R. (1997). Word translation at three levels
of proficiency in a second language: The ubiquitous involvement
of conceptual memory. Language Learning, 47, 215–264.
De Groot, A. M. B., Dannenburg, L., & Van Hell, J. G. (1994).
Forward and backward word translation by bilinguals. Journal
of Memory and Language, 33, 600–629.
Gee, N. R., Nelson, D. L., & Krawczyk, D. (1999). Is the concreteness
effect a result of underlying network connectivity? Journal of
Memory and Language, 40, 479–497.
Gollan, T. H., & Acenas, L. A. (2004). What is a TOT? Cognate and
translation effects on tip-of-the-tongue states in Spanish–English
and Tagalog–English bilinguals. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 30, 246–269. doi:10.1037/
0278-7393.30.1.246
Gollan, T. H., & Silverberg, N. B. (2001). Tip-of-the-tongue states in
Hebrew–English bilinguals. Bilingualism: Language and Cognition,
4, 63–83. doi:10.1017/S136672890100013X
Gollan, T. H., Montoya, R. I., Fennema-Notestine, C., & Morris, S. H.
(2005). Bilingualism affects picture naming but not picture classification.
Memory & Cognition, 33, 1220–1234.
Grondin, R., Lupker, S. J., & McRae, K. (2009). Shared features
dominate semantic richness effects for concrete concepts. Journal
of Memory and Language, 60, 1–19.
Gupta, P., Lipinski, J., Abbs, B., Lin, P.-H., Aktunc, E., Ludden, D., . . .
Newman, R. (2004). Space aliens and nonwords: Stimuli for
investigating the learning of novel word–meaning pairs. Behavior
Research Methods, Instruments, & Computers, 36, 599–603.
doi:10.3758/BF03206540
Hamilton, M., & Rajaram, S. (2001). The concreteness effect in implicit
and explicit memory tests. Journal of Memory and Language, 44,
96–117.
Jin, Y.-S. (1990). Effects of concreteness on cross-language priming in
lexical decision. Perceptual and Motor Skills, 70, 1139–1154.
Kaushanskaya, M. (in press). Cognitive mechanisms of word learning
in bilingual and monolingual adults: The role of phonological
memory. Bilingualism: Language and Cognition. doi:10.1017/
S1366728911000472
Kaushanskaya, M., & Marian, V. (2009). The bilingual advantage in
novel word learning. Psychonomic Bulletin & Review, 16, 705–710.
Kaushanskaya, M., & Rechtzigel, K. (2012). What can errors tell us
about differences between monolingual and bilingual word learning?
Manuscript submitted for publication.
Kroll, J. F., & Stewart, E. (1994). Category interference in translation
and picture naming: Evidence for asymmetric connections between
bilingual memory representations. Journal of Memory
and Language, 33, 149–174.
Marian, V., Blumenfeld, H. K., & Kaushanskaya, M. (2007). The
Language Experience and Proficiency Questionnaire (LEAP-Q):
Assessing language profiles in bilinguals and multilinguals. Journal
of Speech, Language, and Hearing Research, 50, 940–967.
Miller, L. M., & Roodenrys, S. (2009). The interaction of word
frequency and concreteness in immediate serial recall. Memory
& Cognition, 37, 850–865.
Paivio, A. (1986). Mental representations. Oxford, U.K.: Oxford University
Press.
Paivio, A. (1991). Dual coding theory: Retrospect and current status.
Canadian Journal of Psychology, 45, 255–287.
Paivio, A., & Desrochers, A. (1980). A dual-coding approach to
bilingual memory. Canadian Journal of Psychology, 34, 388–399.
Paivio, A., & Yuille, J. C. (1969). Changes in associative strategies and
paired-associate learning over trials as a function of word imagery
and type of learning set. Journal of Experimental Psychology, 79
(3, Pt. 1), 458–463. doi:10.1037/h0026929
Paivio, A., Walsh, M., & Bons, T. (1994). Concreteness effects on
memory: When and why? Journal of Experimental Psychology:
Learning, Memory, and Cognition, 20, 1196–1204.
Papagno, C., & Vallar, G. (1995). Verbal short-term memory and
vocabulary learning in polyglots. Quarterly Journal of Experimental
Psychology, 48A, 98–107.
Romani, C., McAlpine, S., & Martin, R. C. (2008). Concreteness effects
in different tasks: Implications for models of short-term memory.
Quarterly Journal of Experimental Psychology, 61, 292–323.
Schwanenflugel, P. J., & Shoben, E. J. (1983). Differential context
effects in the comprehension of abstract and concrete materials.
Journal of Experimental Psychology: Learning, Memory, and
Cognition, 9, 82–102.
Tokowicz, N., & Kroll, J. F. (2007). Number of meanings and concreteness:
Consequences of ambiguity within and across languages.
Language and Cognitive Process, 22, 727–779. doi:10.1080/
01690960601057068
Van Hell, J. G., & De Groot, A. M. B. (1998). Conceptual representation
in bilingual memory: Effects of concreteness and cognate
status in word association. Bilingualism: Language and Cognition,
1, 193–211.
Van Hell, J. G., & Mahn, A. C. (1997). Keyword mnemonics versus
rote rehearsal: Learning concrete and abstract foreign words by
experienced and inexperienced learners. Language Learning, 47,
507–546.
Vitevitch, M. S., & Luce, P. A. (2004). A Web-based interface to
calculate phonotactic probability for words and nonwords in
English. Behavior Research Methods, Instruments, and Computers,
36, 481–487. doi:10.3758/BF03195594
Walker, I., & Hulme, C. (1999). Concrete words are easier to recall
than abstract words: Evidence for a semantic contribution to
short-term serial recall. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 25, 1256–1271.
Psychon Bull Rev (2012) 19:935–941 941
Copyright of Psychonomic Bulletin & Review is the property of Springer Science & Business Media B.V. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s
express written permission. However, users may print, download, or email articles for individual use.
